On the origins of nutritional slogans I: Saturated fat

Eat less saturated fat. This slogan, along with others like it, has been pressed on us by nutritional authorities over the past 50 years or so. At a population level we largely listened and modified our behaviour. Meanwhile, obesity and other ‘diseases of civilisation’ steadily increased. Objectively, the message has been a failure and, disturbingly, it may have done us harm. The science does not support the message.

Recently, I posted on how nutritional science usually works. If you read it, you might have noticed that the definitive science can only come about after decades of research. These are the stage 3 prospective controlled studies that are not started until late in the scientific process. What that means is that nutritional slogans, like the eat less saturated fat one, cannot be based on solid science at the time they are proclaimed. At best they are plausible hunches.

I thought it worthwhile to put some of these slogans into perspective by examining how they got started and how they held on, starting with saturated fat.


ancel_keysDr Ancel Keys (1904-2004) was an American epidemiologist who was interested in the effects of diet on health. During world war II, Keys developed the K-ration given to soldiers in the field (the ‘K’ was no coincidence), and studied starvation and recovery in conscientious objectors (at a time before ethics committees). He was described as ‘confident’.

Key’s starvation work led him to travel to the post-war Mediterranean where heart disease appeared to have declined with reduced food availability. He ended up concluding that it was a reduction in animal fats that gave protection from heart disease and started actively promoting what would evolve into the ‘Mediterranean Diet’. He didn’t just think that animal fats might be associated with heart disease, he considered them a cause and proposed the link – cholesterol. He compiled heart disease and saturated fat consumption data from countries in northern and southern Europe, the USA and Japan and used it to campaign in support of his conviction:

AncelKeysSelect

It looked convincing: higher fat = higher mortality from heart disease (CHD). However, the problem is that there was data available to him from another 22 countries, not just these ones. He left that bit out. Here’s what the plot would have looked like if he had included all the available data at the time:

AncelKeysAll

There is still a relationship, but not as impressive. When newer data gets added in, it is all over the place.

This approach has dangers in interpretation. Here are some data showing a link between cheese consumption and death from getting entangled in one’s bedsheets (from a site called spurious correlations – the data is real though):

chart

This is not to discredit studies by association. However, they do need to be interpreted cautiously, and it’s usually fairly easy to come up with a plausible explanation for even the weirdest associations – e.g. eating cheese gives people nightmares that cause them to thrash about and get entangled in their bedsheets. You would need a stage 3 prospective study to prove that though.

Around the same time, white middle-aged male politicians with large waistlines were getting increasingly alarmed by the sudden rise in heart disease in the American population. They wanted someone to tell them what to do about it.

Although Keys was claiming he had the ‘key’, his work was criticised by fellow scientists from the outset (not that that meant he was wrong). Preliminary studies were not showing a benefit in low fat diets, and they worried that fat was an essential macronutrient that the body needed to function properly (e.g. after water, fat makes up 70% of the brain). They cautioned that restricting dietary fat could be dangerous. In 1957, the American Heart Association (AHA) refused to endorse reducing saturated fat in the American diet.

Then, in 1961, the AHA did an about-face and recommended a low fat-high carbohydrate diet for Americans at risk of heart disease (not for the general public).

images-19Two things had changed. First, a study (the Framingham study) was published providing evidence that blood cholesterol was a risk for heart disease (along with blood pressure and ECG abnormalities). But, the second was perhaps more important: Ancel Keys was elected to the board of the AHA. The two events got Keys on the cover of Time magazine.

The fate of fat was settled, it had met its nemesis. The AHA bowed to political pressure to do something about the rise in heart disease, Keys was a persuasive and charismatic individual now with authority, there was a study that appeared on-message (although it had not shown that fat was a risk for heart disease, just cholesterol from unknown causes) and finally, the message was nice and simple and the popular press welcomed it.

So that’s how it all started. Biased analysis by a charismatic individual, influence, political pressure, a bit of basic science and media support set up a nutritional message that would persist for the next 50 years.

Nevermind that there was no hard science to support it – the scientists would surely find that quickly enough.

The authorities were in for a surprise. A brief history:

1961

photo.jpgThe AHA recommends that people at risk of heart disease limit total fat intake to 30% of calories (10% saturated; 20% unsaturated). To keep some perspective, Keys himself only recommended limiting carving meats (roasts, chops etc) to 3 times a week.

1969

The National Institutes of Health sets up a Diet Review Panel of experts in epidemiology, statistics, nutrition and clinical medicine to review the evidence for the hypothesis that saturated fat increases the risk of heart disease. They find that the available evidence did not support the hypothesis and, furthermore, express concerns that reducing fat intake could have adverse consequences.

1976

A Select Committee on Nutrition and Human Needs ponders the recommendation for dietary fat to be included in the upcoming ‘US Dietary Goals’. This highly-contentious committee, comprising mostly non-experts in the field and written by a vegan junior staffer without a scientific background, recommends extending the AHA guidelines (until then for people at risk of heart disease) to the entire American adult public.

This causes a furor in the scientific community. Even the American Medical Association protests.

nas_logo_0711Subsequent campaigns by scientists to have the recommendation overturned put the scientists apparently on the same side as the dairy and beef industries and so discredit them by association. Even an organisation as prestigious as the National Academy of Sciences doesn’t escape. The Academy directs its Food and Nutrition Board to provide a science-based assessment of the current position. The Board concludes: just watch your weight – everything else will take care of itself.

In the nutritional quagmire that we find ourselves in today, this could well be the best advice ever issued.

The authorities don’t take the Board’s conclusion well. Not only does it reject their recommendation, it undermines their own relevance.  The Board becomes the subject of a determined public campaign accusing it of being an industry apologist and having industry links. The media frenzy eventually succeeds in obscuring the science and discrediting the Board and its findings.

The US Department of Agriculture (USDA), the government body with the last word on the matter, then takes up the saturated fat message and begins putting the recommendations into practice throughout the US.

1980-1984

A series of well-controlled saturated fat studies are published that consistently find no link to heart disease.

1984

Time-Magazine-CholesterolThe results from the Coronary Primary Prevention Trial are published, confirming the Framingham study by demonstrating a link between blood levels of cholesterol and heart disease. They then do an amazing thing – they conclude that this also applies to diet as well. This triggers a renewed public campaign, another cover on Time magazine, and the extension of the AHA guidelines to every American over the age of 2 years (regardless of anything else).

1988

The US Surgeon General’s Office commissions what it hopes will be the definitive report linking saturated fat to heart disease. The task is to compile the science and get it reviewed by an expert committee who would issue a proclamation. The committee is tasked with supporting the hypothesis that saturated fat increases the risk of heart disease (it is not required to compare the scientific data for or against the hypothesis). It takes longer than expected (see 1999).

1990

A disturbing turn of events. Data starts to appear showing that low levels of dietary saturated fat might be associated with other mortalities such as cancer and respiratory illness. A troubling finding was an increased likelihood of death by accident and trauma – the brain might have gone into a high-risk survival mode. The fears that dietary advice could do harm were realised – not surprisingly, fat had a bigger role to play in our biology than just the heart. The authorities do nothing – they ignore the most fundamental principle in medicine – do no harm.

1999

It has now been eleven years since the US Surgeon General’s Office committee was set up (in 1988). They give up. In the face of the complexity of the task and the simple fact that the science was not supporting the hypothesis, the committee quietly admits defeat and disbands without fanfare.

2000-2010

Large meta-analyses of prospective (stage 3) studies appear in the following decade. The conclusion remains the same – despite decades of research effort and who knows how many hundreds of millions in precious research dollars, it is still not possible to definitively show that saturated fat has any significant part to play in heart disease.

This should come as no surprise really. Given the diet that our species evolved with, it is highly unlikely that our biology would have penalised us for eating animal fat.

2015

Today, the dietary advice to avoid saturated fat remains unchanged. The authorities refuse to change the message without what they call convincing proof (as if they don’t have that already, in spades). As the Harvard epidemiologist W. Willett points out, the irony is that they didn’t bother with convincing proof when they set their guidelines in the first place.


Without scientific backing, how is it that the message has prevailed? The message was readily taken up by health professionals, government agencies, health advocates and nearly everyone else. It would be a brave health professional who strayed from the message publicly. The media has been particularly receptive and helpful to the cause. A low fat industry that didn’t exist before has flourished and is profitable. Middle-aged people living today have never heard a different message. Its truth is now taken for granted – the science doesn’t have a voice.

But what if it is true after all and decades of painstaking scientific research is wrong? What if saturated fat raises blood cholesterol that in turn increases the risk of heart disease? What about the lives lost if we get this wrong?

First: lives may have been shortened by this advice. Low levels of dietary saturated fat might be associated with various mortalities (see above). The compensatory increase in carbohydrate consumption (simple and complex) has probably increased body-mass (another risk factor for heart disease and type II diabetes etc). We were urged to switch from butter to margarine, meaning that whole populations have been eating trans-fats for decades (including children). The unsaturated vegetable oils that we have been using contain excessive omega-6 (an essential fatty acid, but pro-inflammatory in excess), and are prone to the development of free radicals and trans-fats with repeated heating. Whether this has real adverse health consequences I cannot say. Equally, it cannot be assumed that the eat less saturated fat message does no harm.

Second: Often nutritional authorities talk about lives saved or deaths avoided. As if you can avoid death. What they should say is deaths delayed. And then they should estimate how delayed. This has been done, based on the hypothetical benefit of reducing saturated fat assuming saturated fat was as important as claimed. The result? A lifetime of ’taking the skin off the chook’ and other sad deprivations might delay death by a few days to at most a few (3-4) months.

In a nutshell

Our diet has been experimented on at a national level with an hypothesis that started out as a campaign by a charismatic and influential individual, that went on to become lore, that didn’t and doesn’t have scientific backing, that even if right would have a negligible effect on our lifespan and that could possibly be harmful to our health.


Print This Post Print This Post

Further reading:

The Gary Taubes article was highly influential when published in 2001: The soft science of dietary fat

An update by J Elliott in 2014: Flaws, fallacies and facts: Reviewing the early history of the lipid hypothesis

Just this year (2015) the British Medical Journal (BMJ) published a blistering editorial on the fallacy of the saturated fat guidelines.

Scientists are now questioning the value of nutritional advice and the minimum level of evidence needed before issuing it: A call for higher standards of evidence for nutritional guidelines.

Related posts on this site:

Heritage fats: http://www.6xc.com.au/animal-fats/

Fat at a molecular level: http://www.6xc.com.au/beginners-guide-fat/

Fish oil and omega-3/6: http://www.6xc.com.au/fish-oil/